A relative evaluation of multiclass image classification by support vector machines

Support vector machines (SVMs) have considerable potential as classifiers of remotely sensed data. A constraint on their application in remote sensing has been their binary nature, requiring multiclass classifications to be based upon a large number of binary analyses. Here, an approach for multiclass classification of airborne sensor data by a single SVM analysis is evaluated against a series of classifiers that are widely used in remote sensing, with particular regard to the effect of training set size on classification accuracy. In addition to the SVM, the same datasets were classified using a discriminant analysis, decision tree, and multilayer perceptron neural network. The accuracy statements of the classifications derived from the different classifiers were compared in a statistically rigorous fashion that accommodated for the related nature of the samples used in the analyses. For each classification technique, accuracy was positively related with the size of the training set. In general, the most accurate classifications were derived from the SVM approach, and with the largest training set the SVM classification was significantly (p < 0.05)more accurate (93.75%) than that derived from the discriminant analysis (90.00%) and decision tree algorithms (90.31%). Although each classifier could yield a very accurate classification, > 90% correct, the classifiers differed in the ability to correctly label individual cases and so may be suitable candidates for an ensemble-based approach to classification.

[1]  E Comak,et al.  A support vector machine using the lazy learning approach for multi-class classification , 2006, Journal of medical engineering & technology.

[2]  Andreu Català,et al.  K-SVCR. A support vector machine for multi-class classification , 2003, Neurocomputing.

[3]  G. H. Rosenfield,et al.  A coefficient of agreement as a measure of thematic classification accuracy. , 1986 .

[4]  Giles M. Foody,et al.  The effect of training set size and composition on artificial neural network classification , 1995 .

[5]  Chih-Jen Lin,et al.  A comparison of methods for multiclass support vector machines , 2002, IEEE Trans. Neural Networks.

[6]  Mark A. Friedl,et al.  Using prior probabilities in decision-tree classification of remotely sensed data , 2002 .

[7]  C. Brodley,et al.  Decision tree classification of land cover from remotely sensed data , 1997 .

[8]  Dustin Boswell,et al.  Introduction to Support Vector Machines , 2002 .

[9]  Steven E. Franklin,et al.  Multisource evidential classification of surface cover and frozen ground , 1992 .

[10]  G. Mercier,et al.  Support vector machines for hyperspectral image classification with spectral-based kernels , 2003, IGARSS 2003. 2003 IEEE International Geoscience and Remote Sensing Symposium. Proceedings (IEEE Cat. No.03CH37477).

[11]  S. Prasher,et al.  Classification of hyperspectral data by decision trees and artificial neural networks to identify weed stress and nitrogen status of corn , 2003 .

[12]  Jon Atli Benediktsson,et al.  The effect of classifier agreement on the accuracy of the combined classifier in decision level fusion , 2001, IEEE Trans. Geosci. Remote. Sens..

[13]  D. Peddle,et al.  Multi-Source Image Classification II: An Empirical Comparison of Evidential Reasoning and Neural Network Approaches , 1994 .

[14]  Giles M. Foody,et al.  Log-linear modelling for the evaluation of the variables affecting the accuracy of probabilistic, fuzzy and neural network classifications , 1997 .

[15]  A. Belousov,et al.  A flexible classification approach with optimal generalisation performance: support vector machines , 2002 .

[16]  Martin Brown,et al.  Linear spectral mixture models and support vector machines for remote sensing , 2000, IEEE Trans. Geosci. Remote. Sens..

[17]  Nello Cristianini,et al.  An introduction to Support Vector Machines , 2000 .

[18]  Stephen V. Stehman,et al.  Selecting and interpreting measures of thematic classification accuracy , 1997 .

[19]  Yue Wang,et al.  Sphere-Structured Support Vector Machines for Multi-class Pattern Recognition , 2003, RSFDGrC.

[20]  Giles M. Foody,et al.  Status of land cover classification accuracy assessment , 2002 .

[21]  G. Foody.  Thematic map comparison: Evaluating the statistical significance of differences in classification accuracy , 2004 .

[22]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[23]  I. Kanellopoulos,et al.  Strategies and best practice for neural network image classification , 1997 .

[24]  Miguel Figueroa,et al.  Competitive learning with floating-gate circuits , 2002, IEEE Trans. Neural Networks.

[25]  Fabio Roli,et al.  Design of effective neural network ensembles for image classification purposes , 2001, Image Vis. Comput..

[26]  Martin Brown,et al.  Support vector machines for optimal classification and spectral unmixing , 1999 .

[27]  G. G. Wilkinson,et al.  Evidential reasoning in a pixel classification hierarchy—a potential method for integrating image classifiers and expert system rules based on geographic context , 1990 .

[28]  Johannes R. Sveinsson,et al.  Multiple classifiers applied to multisource remote sensing data , 2002, IEEE Trans. Geosci. Remote. Sens..

[29]  D. Roberts,et al.  A comparison of methods for monitoring multitemporal vegetation change using Thematic Mapper imagery , 2002 .

[30]  L. Zhang,et al.  Comparison of Neural Networks and Statistical Methods in Classification of Ecological Habitats Using FIA Data , 2003 .

[31]  P. Swain,et al.  Neural Network Approaches Versus Statistical Methods In Classification Of Multisource Remote Sensing Data , 1990 .

[32]  Donato Malerba,et al.  A Comparative Analysis of Methods for Pruning Decision Trees , 1997, IEEE Trans. Pattern Anal. Mach. Intell..

[33]  Fernando Pérez-Cruz,et al.  Puncturing Multi-class Support Vector Machines , 2002, ICANN.

[34]  Ethem Alpaydin,et al.  Support Vector Machine for Multiclass Classification , 1998 .

[35]  Giles M. Foody,et al.  The significance of border training patterns in classification by a feedforward neural network using back propagation learning , 1999 .

[36]  Giles M. Foody,et al.  On the compensation for chance agreement in image classification accuracy assessment, Photogram , 1992 .

[37]  Bernard A. Engel,et al.  Optimization of training data required for neuro-classification , 1994 .

[38]  J. Anthony Gualtieri,et al.  Support vector machines for hyperspectral remote sensing classification , 1999, Other Conferences.

[39]  Johannes R. Sveinsson,et al.  Support vector machines in multisource classification , 2003, IGARSS 2003. 2003 IEEE International Geoscience and Remote Sensing Symposium. Proceedings (IEEE Cat. No.03CH37477).

[40]  Yoonkyung Lee,et al.  Classification of Multiple Cancer Types by Multicategory Support Vector Machines Using Gene Expression Data , 2003, Bioinform..

[41]  P. A. Blamire.  The influence of relative sample size in training artificial neural networks , 1996 .

[42]  Paul M. Mather,et al.  An assessment of the effectiveness of decision tree methods for land cover classification , 2003 .

[43]  L. S. Davis,et al.  An assessment of support vector machines for land cover classi(cid:142) cation , 2002 .

[44]  A. Jawad,et al.  Computer processing of remotely sensed images , 2005 .

[45]  Steven A. Ackerman,et al.  Cloud Classification of Satellite Radiance Data by Multicategory Support Vector Machines , 2004, Journal of Atmospheric and Oceanic Technology.